Skip to content

Latest commit

 

History

History
59 lines (46 loc) · 3.35 KB

README.md

File metadata and controls

59 lines (46 loc) · 3.35 KB
Raw

The following steps show how to prepare training dataset to train the mode.

Libraries to install

    pip install ftfy langdetect numpy torch pandas nltk sentencepiece boto3 tqdm regex bs4 newspaper3k htmlmin tldextract 
    git clone https://github.com/mattilyra/LSH
    cd LSH
    python setup.py install

Download the dataset

  1. Download the deduplicated URLs from jcpeterson
  2. Remove blacklisted URLs.
python blacklist_urls.py <path to the downloaded deduplicated URLs> <filename for clean urls. e.g. clean_urls.txt>

  1. Download the content from the clean urls with openwebtext's utilities.

  2. Merge the contents into one loose json file with 1 json per newline of the format {'text': text, 'url': unique_url}. It is important for the url to be unique.

Prepare the data for GPT training:

  1. Perform ftfy, english detection and remove documents with less than 128 tokens. This step can be sharded and run on shards.
python cleanup_dataset.py <input data file> <output cleaned data filename>

Additional cleanup (e.g. remove documents less than 512 characters or dataset specific cleaning like stories, realnews datasets) can be done using cleanup_fix_dataset.py. More details can be found by running python cleanup_fix_dataset.py --help. 2. Using LSH, find possible duplicates and store then in a file for later processing. The code supports saving and loading fingerprints for recurrent deduplications, and is also multithreaded for faster processing. More details are can be found by python find_duplicate.py --help.

python find_duplicates.py --inputs <pairlist list of input cleaned data files and keys, e.g. cc.json cc_id news.json news_id> --output <output possible duplicate urls filename>
  1. Based on similarity measure defind inside function is_similar (default: 0.9), group urls that are similar. Basically, for each group, only one url we should keep and remove the rest.
python group_duplicate_urls.py <possible duplicate urls file> <output file containing similar urls>
  1. Remove similar documents that were detected in the last step.
python remove_group_duplicates.py <file containing similar documents> <cleaned data file> <outputfile containing deduplicate data>
  1. Shuffle the dataset.
shuf <cleaned deduped data file> -o train_data.json

Deduplicating ngrams

To deduplicate the downstream tasks (e.g. lambada, squad) from the training dataset, we run the following command.

python filter_ngrams.py --tasks <name of the task, e.g. lambada, squad> --dedup-dataset <training dataset to deduplicate> <json key> --output <output training dataset>

We use 13-grams by default for the deduplication. When we find a 13-gram match in a training document, we split the document into two pieces and remove the 13-gram along with 200 characters from the both side of the 13-gram. We also remove any splitted document with less than 200 characters or if a document got splitted more than 10 times. These parameters can be changed using corresponding arguments.

Only for the lambada task, we need to provide the path, --lambada-path <path of the lambada test data>.

Several other features (e.g. save and load dictionary) have been added, look at python filter_ngrams.py --help for details.